Skip to main content

9- by 15-Foot Low Speed Wind Tunnel Acoustic Improvements Expanded Overview

GRC-E-DAA-TN31946 · NASA (NTRS) · 2016

Public domain · NASA (NTRS)Technical Reports

Overview

The 9- by 15-Foot Low Speed Wind Tunnel (9x15 LSWT) at NASA Glenn Research Center was built in 1969 in the return leg of the 8- by 6-Foot Supersonic Wind Tunnel (8x6 SWT). The 8x6 SWT was completed in 1949 and acoustically treated to mitigate community noise issues in 1950. This treatment included…

Publisher
NASA (NTRS)
Document
GRC-E-DAA-TN31946
Year
2016
Pages
35

Document

9 - by 15 - Foot Low Speed Wind Tunnel

Acoustic Improvements

Expanded Overview

Acoustics Branch Point - of - Contact: David Stephens

April 2016

Executive Summary

The 9 - by 15 - Foot Low Speed Wind Tunnel (9x15 LSWT) at NASA Glenn

Research Center was built in 1969 in the return leg of the 8 - by 6 - Foot

Supersonic Wind Tunnel (8x6 SWT). The 8x6 SWT was completed in

1949 and acoustically treated to mitigate community noise issues in

1950. This treatment included the addition of a large muffler

downstream of the 8x6 SWT test section and diffuser.

The 9x15 LSWT was designed for performance testing of V/STOL

aircraft models, but with the addition of the current acoustic

treatment in 1986 the tunnel been used principally for acoustic and

performance testing of aircraft propulsion systems. The present

document describes an anticipated acoustic upgrade to be completed

in 2017 .

8x6/9x15 Facility Manager : David Stark

Unique facility for testing propulsors

9x15 LSWT 8x6 SWT Mach 0 to 0.23 0 to 0.1 Number 0.25 to 2.0 Simulated Sea Level 1000 to 35,000 ft Altitude

Fan/Propulsor Testing in 9x15 Tunnel

• The GRC 9x15 Low Speed Wind Tunnel has

Honeywell Turbofan with Acoustic Liners

been extensively used to study and acoustically

characterize nearly all of the NASA/Industry

propulsor concepts over the past 20 years.

• Except for maintenance, the acoustic

treatment has remained essentially unchanged

in 20+ years.

NASA/GE Open Rotor P&W Geared Turbofan P&W Advanced Ducted Propulsor

Testing quiet fans requires

a quiet wind tunnel

• Historically this has been done by testing

a Mach 0.1, which is below true take - off

and landing speeds.

Empty 9x15 vs Low Power ADP with Liners Empty 9x15 vs Low Power ADP with Liners Empty 9x15 vs Low Power ADP with Liners

• Future fans may be even

90 90

quieter

80 80 80

• Low tip speed

• Low pressure ratio

70 70 70

• Acoustic liners

• Open rotors and other

60 60 60 ADP Aft (126º emitted)

Third Octave SPL, dB concept fans require

Third Octave SPL, dB Third Octave SPL, dB ADP Aft (126º emitted) - - ADP Aft (126º emitted) - ADP Broadside (89º emitted)

testing at higher tunnel

ADP Broadside (89º emitted) ADP Broadside (89º emitted) One 50 One 50 One 50 ADP Forward (23º emitted)

speeds than Mach 0.1

ADP Forward (23º emitted) Current Empty 9x15, Mach 0.10 ADP Forward (23º emitted) Current Empty 9x15, Mach 0.10 40 40 Current Empty 9x15, Mach 0.20 100 1000 10000 100000 100 1000 10000 100000 100 1000 10000 100000 Frequency, Hz Frequency, Hz Frequency, Hz

Support from NASA Agency and Center Levels

• Preliminary assessment by Jacobs Technology, Inc performed in 2012

• Funded by Environmentally Responsible Aviation

• Initial FY15 NASA funding resulted in initiation of 9x15 Design

contract that resulted in completion of a 30% design review.

• Additional funding in FY15 from NASA GRC, NASA AETC project and

NASA Augmentation funded a 9x15 Design/Build contract which

began in September 2015 from the 30% design point. The FY15

funding provided 60% of the total project Design/Build cost.

• FY16 funding from NASA GRC and NASA AETC is currently being

approved by Congress through the FY16 NASA Operating Plan and

this funding will provide the remaining 9x15 Design/Build funding .

Anticipated Schedule

• December 2015 – Final pre - construction aerodynamic calibration of

8x6 and 9x15 test sections

• January 2016 – Final pre - construction background noise

measurement of 9x15 test section

• September 30, 2016 – Boundary layer ingestion test in 8x6 tunnel

ends

• October 31, 2016 – Wind Tunnel Shutdown for Start of Site Work

• Both 8x6 and 9x15 shut down

• September 4, 2017 – Acceptance Testing Begins

• Acceptance testing for 8x6 and 9x15 test sections

• September 29, 2017 – Tunnel Ready for Testing

Planned Wind Tunnel Renovation

Complementary but discrete improvements

1. Add fairings and turning vanes to turn 2

2. Add acoustic baffles downstream of doors 1 & 2

3. Replace test section flow surfaces, remove slots

4. Reshape diffuser and add acoustic treatment

5. Add turning vanes to turn 3

1 2 5

Flow

9x15 Tunnel, Top View

Turn 2 Turning Vanes

Cooler

Muffler

Exit

Improvements to Turn 2

Velocity Field in Turn 2

Heat Muffler Exchanger Exit

Case 1: Existing design

Case 2 : Proposed C - 2 design

Proposed designs provides much

more uniform airflow into cooler.

This should result in more uniform

temperature in the test section.

Downstream of Cooler

Looking upstream at the Cooler

Test Section

Looking Downstream from Inlet

Current 9x15

Test Section

• The test section surface is

perforated steel over a bulk

Kevlar absorber

• Facing plate is 16 ga , 1/8”

holes, 40 percent open

• The acoustic treatment is

built as dozens of individual

boxes

• There are slots in both walls,

and many seams throughout

the tunnel

• The 2012 study by Jacobs

concluded that the majority

of the noise in the 9x15 test

section above 2 kHz is due to

boundary layer flow over

perforated steel surface

Test Section Perforated Metal (Current)

Jacobs concluded that the noise in

From Summer

the 9x15 test section above 2 kHz is

due to airflow over the perforated

2012 Study

steel tunnel walls

90 90 Prediction:

~ 7 dB reduction

80 80 70 70 60 60 50 50 Measured In-Flow SPL Measured In-Flow SPL NASA 9x15 SPL Goals in Test Section NASA 9x15 SPL Goals in Test Section 40 40 BLNoise Estimate (roughness = 1.5mm) BLNoise Estimate (roughness = 1.5mm) BLNoise Estimate (roughness = 0.1mm) 1/3rd Octave Band Sound Pressure Levels (dB) 1/3rd Octave Band Sound Pressure Levels (dB) 30 30 100 1000 10000 100000 100 1000 10000 100000 Frequency (Hz) Frequency (Hz) Front

Proposal to improve

the 9x15

• Jacobs estimated a 7 dB

reduction in roughness noise

is possible by replacing the

current wind tunnel surface

with a woven glass fiber

cloth, lowering roughness

from 1.5mm to 0.1mm (not

including seams)

• It is a synthetic fabric made of

flat weave bonded to a

coated perforated metal

sheet

• Sample shown 

Back

Roughness Noise Facility

at Virginia Tech

Inside Virginia Tech Facility 2014

Test section modification

successful, allowing

flush mounted samples

with deep acoustic

treatment below.

Extensive roughness noise testing at Virginia

Tech to find quiet surface

• More than 30 samples tested

Substantial noise benefit possible by

• Combinations of

replacing current test section flow surface • Perforate (hole size, % open, thickness) • Covering (glass cloth, wire mesh)

10 dB

PSD #6 PSD,

-10 -5

9x15 Baseline Mic -10 Solid Plate

-20 -15

5/32 63% 200x600 -20 0 0.5 1 1.5 2 2.5 Frequency, Hz x 10

-30

0 5000 10000 15000 20000

• Very wide range of results

Frequency, Hz

• For more information, see AIAA - 2015 - 3261 21

Diffusion Bonded Plates

Diffusion bonded plates (DBP) are produced by combining perforated sheet metal with fine

wire cloth. These materials are joined in a vacuum furnace under heat and mechanical

load in a process called diffusion bonding.

The function of the DBP is to allow passage of sound waves into bulk absorber material, while producing minimal self - noise under grazing flow.

DBP Details:

• 5/32” holes on 3/16” centers

(63% open area) 16 - gauge

perforated sheet metal

• 200x600 Twilled Dutch Weave

wire cloth

• 304 stainless steel cloth and

plate

• Wire cloth “dimples”

(depressions over perforations)

approx. 0.005” deep

• Flow resistivity ~12 CGS Rayls

Goal: Minimal impact to anechoic quality

• Lower self noise AND maintain anechoic quality

• Impact of wire cloth over perforate has been investigated,

examples given in next four slides

• LaRC Normal Incidence Tube

• Riverbank Acoustical Laboratory

• Glenn Acoustical Testing Laboratory

• LaRC Curved Duct Test Rig

• Testing of reinforced panels in GRC Acoustics Lab is pending

• Anechoic quality of test section to be assessed by external

contractor according to ISO 26101 during July 2016

NASA Langley Normal

Incidence Tube

• Established facility

• Built for liner testing

• Limited to 3000 Hz, plane waves

• Ripples due to depth of bulk absorber

1.00 0.95 0.90 0.85 Absorption Coefficient Current 9x15 Perforate Proposed DBP

Langley Normal

0.80

Incidence Tube

0 500 1000 1500 2000 2500 3000 3500 Frequency, Hz

Reverberation room test at Riverbank

Acoustical Laboratory

Six configurations were tested, with variation in bulk absorber density, with and without

perforated panel covering and with different panel rib arrangements.

Reverberation Room Test Results 1.10 1.00 0.90 0.80 Normalized Acoustic Absorptivity higher density composite higher density composite + DBP higher density composite + DBP + solid ribs higher density composite + DBP + perf ribs 0.70 10 100 1000 10000 1/3 OB Frequency, Hz The DBP causes < 3% absorption reduction above 400 Hz.

Reflection Test at NASA Glenn

Acoustical Testing Laboratory

Pressure, Pa -5

• Not limited in frequency range

-10 0 0.005 0.01 0.015 0.02

• Accurate positioning a challenge

Time, Seconds ATL Reflection Test Results 0.98 0.96 0.94 0.92 0.9 0.88 Current 9x15 0.86 Proposed Panel Absorption Coefficient 0.84 10 per. Mov. Avg. (Current 9x15) 0.82 10 per. Mov. Avg. (Proposed Panel) 0.8 0 1000 2000 3000 4000 5000 6000 7000 8000 Frequency, Hz Attenuation of 9x15 Samples Measured at CDTR, Mach 0.20 BaselineM02D00 BaselineM02D01

NASA Langley

BaselineM02D30 WireClothM02D00 WireClothM02D01

Curved Duct Test Rig

WireClothM02D30

• Engine liner test with flow

Insertion Loss, dB

• Rectangular duct modes

• Limited to 3000 Hz

0 500 1000 1500 2000 2500 3000 Frequency, Hz

Diffuser

Looking upstream into the diffuser

Recirculation

Existing

Attached Flow

Modified

Test Section to be lengthened 5 - feet into diffuser

Red Lines denote limits of measurements in upstream and downstream directions. First or last measurement repeated as needed to complete analysis.

• Current test section

length restricts aft

measurement angles

• Floor mounted

microphones are a

marginal solution

• This can have a

significant impact on

EPNL calculations

• The addition of a 5 - foot

straight extension into

the current diffuser will

enable measurement

to 150º geometric from

upstream

Anticipated Improvement to 9x15 noise levels

after all 5 upgrades

90.0 Empty Tunnel, Jacobs 80.0 Frequency (2012) Prediction 630 81.4 65.6 794 69.3 80.8 1000 71.9 80.6 70.0 1260 82.5 73.3 1587 80.8 73.7 2000 78.7 73.9 60.0 2520 79.7 73.1 Third Octave SPL, dB - 3175 81.5 73.5 4000 83.5 73.3 One 5040 85.1 72.7 Current Empty 9x15, Mach 0.20 50.0 6350 85.3 72.4 8000 70.5 84.5 Predicted Empty 9x15, Mach 0.20 10079 68.3 83.1 12699 81.7 64.7 40.0 16000 80.6 63.0 100 1000 10000 100000 20159 79.1 61.3 Frequency, Hz 25398 77.9 58.5 32000 75.1 55.7 Acceptance criteria: No less 40317 71.7 52.6 50797 71.3 49.9 than 3 dB from prediction

Relative to P&W Advanced Ducted Propulsor (ADP)

Model Fan Measurements

Empty 9x15 vs Low Power ADP with Liners Third Octave SPL, dB - ADP Aft (126º emitted) One ADP Broadside (89º emitted) ADP Forward (23º emitted) Current Empty 9x15, Mach 0.20 Predicted Empty 9x15, Mach 0.20 100 1000 10000 100000 Frequency, Hz

Summary

• GRC responding to industry feedback on 9x15 background

noise level requirements

• Design and build contract in place to implement changes to

9x15 for improvements and funding identified and going

through approvals

• The work to date by suggests substantial reductions in

background that will improve signal - to - noise required in

future systems

• Additional work on measurement and signal processing are

expected to create additional signal - to - noise headroom

Additional Materials

Noise Predictions for Other Mach Numbers

Current Predicted 90.0 Mach Mach Mach Mach Mach Mach Frequency 0.10 0.15 0.20 0.10* 0.15* 0.20 80.0 630 62.4 73.1 81.4 53.3 62.1 65.6 5 Pa - 794 62.4 73.1 80.8 55.2 64.4 69.3 1000 62.5 73.6 80.6 57.6 65.2 71.9 70.0 1260 64.3 74.1 82.5 57.7 66.3 73.3 1587 65.7 73.1 80.8 57.8 67.1 73.7 2000 67.1 73.6 78.7 57.0 66.7 73.9 60.0 2520 67.9 75.3 79.7 55.6 67.0 73.1 3175 67.9 76.8 81.5 54.9 66.3 73.5 4000 67.3 77.9 83.5 53.3 65.5 73.3 50.0 Current, Mach 0.20 5040 66.1 78.3 85.1 51.3 64.9 72.7 Third Octave Band SPL, dB ref 2e Current, Mach 0.15 - 6350 63.8 77.3 85.3 46.8 62.9 72.4 Current, Mach 0.10 40.0 One 8000 62.5 75.6 84.5 44.8 60.4 70.5 Predicted, Mach 0.20 10079 60.8 74.4 83.1 42.9 56.5 68.3 *Predicted, Mach 0.15 *Predicted, Mach 0.10 12699 58.2 73.4 81.7 38.8 55.1 64.7 30.0 16000 55.8 71.7 80.6 36.3 53.0 63.0 500 5000 50000 20159 54.1 69.5 79.1 34.9 49.5 61.3 Frequency, Hz 25398 51.7 67.7 77.9 47.1 58.5 32000 48.1 65.0 75.1 44.5 55.7 40317 44.0 61.3 71.7 52.6 50797 42.1 57.3 71.3 49.9 * Provided prediction is for Mach 0.20, scaled to other Mach numbers by NASA. The prediction was scaled in frequency by the Mach number ratio and in amplitude to mimic the current noise.

Source & rights

Source: ntrs.nasa.gov. Public-domain U.S. Government work (17 USC §105) — freely reproducible.

Permanent URL — we don’t break links.

Document details

Doc number
GRC-E-DAA-TN31946
Publisher
NASA (NTRS)
Year
2016
Pages
35
File size
1.5 MB